入侵数据衰减性变化后的优化检测仿真  

The Invasion of Attenuation Change After Optimizing Mining Simulation Data

作　　者：罗刘敏[1] 侯桂云[1] 

机构地区：[1]河南理工大学万方科技学院,河南郑州451400

出　　处：《计算机仿真》2016年第8期234-237,共4页Computer Simulation

摘　　要：在入侵数据优化检测优化的研究中,由于网络入侵特征会随着入侵后时间的推移发生衰减性变化,导致入侵特征发生不断的更新,传统的入侵数据检测方法,对于不断发生特征衰减变化的入侵数据无法固定检测阀值,存在特征识别能力差,检测精度低的问题。提出采用衰减调控的入侵相关数据关联融合挖掘算法。通过获取调控数据集构建入侵相关数据的数据模型,提取入侵相关数据的数据特征并进行预处理。引入多普勒频移模糊搜索对入侵相关数据进行平滑处理,得到入侵相关数据特征包络幅度估计值,对特征提取结果进行衰减调控处理,实现基于衰减调控的入侵相关数据关联融合挖掘。仿真结果表明,改进算法能有效的对入侵相关数据特征进行分类,提高对入侵数据的检测精度。A fusion mining algorithm of intrusion related data with attenuation regulation is proposed. The data model of the intrusion related data is established by acquiring the regulate data set,and the data characteristics of intrusion related data are extracted and preprocessed. The intrusion related data is smoothed by the introduction of the Doppler frequency shift fuzzy search,and the envelope amplitude estimation value of intrusion related data characteristic is obtained. The feature extraction results are made attenuation regulation processing,the association fusion mining of intrusion related data based on attenuation regulation is implemented. The simulation results show that the improved algorithm can effectively categorize the data characteristics related intrusion,and improve the detection accuracy of intrusion data.

关 键 词：入侵 衰减调控 关联融合 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]